Luke Drago: belief

Transcript context

…So I'm biased, but my company seems to be doing a pretty good thing here. And obviously, we're not in stealth. We've announced that we exist. We've got a one-pager of what we're doing, but no one's seen the thing we're working on yet. This fall, we're very excited to roll that out and really show people what we're working on here. But I think there are a couple of categories. We walked through kind of three in the piece. One, and this is kind of counterintuitive, we talk a lot about these kind of defensive acceleration technologies. The idea that you actually have to mitigate AI's catastrophic risks in order to get over this barrier. And the reason for that is because... AI's catastrophic risks provide a very good reason to centralize them in the hands of a couple of people. It is true that by default, AI could be extremely dangerous. It could be extremely powerful and extremely dangerous. It could make it easier for actors to develop bioweapons. It can make it easier for random people to do bad things. And governments are going, and governments and companies are going to use those as credible arguments, real arguments, to centralize its intelligence and de-commoditize it, to have a couple of actors who have dominant control over it. And of course, the downside of that is we know that the more we centralize this into the hands of a couple of people, the more it looks like a monopoly instead of a commodity, the worse off regular people are likely to be in the long run. So what we want to do instead here is de-risk the technology fundamentally. If we're going to build it, and I'm not saying that we do, but if we're going to build it, you should make sure that it's safe. And I think there's been this long-running argument in the AI safety space that doing this is not possible or a waste of time. And we're increasingly seeing interesting results here that indicate maybe actually there's something to be done. Kyle O'Brien had a paper with AC a couple days ago talking about how if you just remove biological materials information from the training data when you do pre-training, that you end up with models that are somewhat tamper resistant, even when you try to reintroduce that later in fine-tuning. That is the kind of research you want to be seeing a whole lot more of right now. You want to find the kind of research that means that if we develop it, doesn't have to be in the hands of 1 actor forever. That one guy has not declared the total controller over intelligence. And then, of course, you really want to work on technology. that helps democratize this tech with humans still in control. Again, part of what we're working on here is trying to find use for these last mile of automation tasks, of taking advantage of an individual's data, finding ways to make that even more competitive for them, even as there are larger models. This sometimes looks like modifying an existing model. It might look like doing something entirely different. finding ways to make that even more competitive for them, even as there are larger models. This sometimes looks like modifying an existing model. It might look like doing something entirely different. But finding ways to put existing human data to use so that the tools that you control are the ones that are helping you do better and that they don't disempower you. also want to work in the kinds of tech that could help strengthen democracies. I think Audrey Tong's kind of vision here is quite inspiring. And so I think those are kind of like the three buckets I talk about. Tech that actually makes it possible so that if we build it, it's going to be diffuse as opposed to a monopoly. Tech that keeps humans firmly in charge. and technology that is able to help strengthen our democracies such that if we can't prevent them from being a monopoly, we have fallback options. One of the ways to think about this to close this is, to close this loop here, is on social media. I think there are two problems in social media, or two approaches. I think you should take them both concurrently. One approach is to say, the kind of common one, is that social media is super addictive, and so the government should regulate it in some way. The government should restrict certain kinds of features that are in it or age-gated or something like this. I think an approach that is oftentimes less appreciated and is absolutely necessary because you can only regulate things so much is to also introduce technological alternatives. There has been a massive rise of like screen time maps, for example, Opal's one of them, where you download a thing and it helps you reclaim your focus because a whole lot of algorithms are pointed at you and now you need something pointed outwards. We're trying to build the thing that's pointed outwards that so many people are trying to take your job or take you out of the economy and we think we can build tools to keep you in it. And I think if we're right, that could be one of the largest markets in history because if you are building the tools that help keep people involved, people are going to want to be involved. They're going to want to stay involved in the future. And I think that's a pretty powerful tool to be building, both from an impact perspective and from a market perspective. We're facing this tension between trying to control the downsides of AI by centralizing it and then spreading the upside by giving as many people as possible access to the models. So one answer to this tension is just to say that we need to open source AI fully. What do you think about that vision and how does it interface with what you're talking about?…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.